[PATCH -perfbook 1/2] memorder: Put \FloatBarrier in front of Section 15.3

[PATCH -perfbook 1/2] memorder: Put \FloatBarrier in front of Section 15.3

[Akira Yokosawa]

Commit e39bbe1b0856 ("memorder: Add example showing store breaking
release-acquire chain") added a couple of code snippets in the
final subsection of Section 15.2.
In 1c builds, this caused the beginning part of Section 15.3 to
appear in the middle of floating code snippets of Section 15.2.

This can be prevented by putting \FloatBarrier in front of Section
15.3, with a (somewhat) wide vertical space left at the end of
Section 15.2.

Note: \FloatBarrier does nothing unless there are some overloaded
float objects. Therefore, 2c and eb builds are not affected.

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
Hi Paul,

Another option would be to use the "section" option of "placeins"
packages as is done for eb builds. I tried but I got quite a few
sparse-looking pages in other part of 1c builds. So I think adding
a \FloatBarrier here is a reasonable thing to do.

        Thanks, Akira
--
 memorder/memorder.tex | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/memorder/memorder.tex b/memorder/memorder.tex
index 66f75662f26c..f85cfe6ca702 100644
--- a/memorder/memorder.tex
+++ b/memorder/memorder.tex
@@ -2404,6 +2404,8 @@ sea of more complex memory-ordering constraints.
 % @@@ Exercises?
 % @@@ Hardware details from Appendix?
 
+\FloatBarrier
+
 \section{Compile-Time Consternation}
 \label{sec:memorder:Compile-Time Consternation}
 %

base-commit: 43d8aa12c84c8b26ffdb8479b4d793cc0159030e
-- 
2.25.1

[PATCH -perfbook 2/2] memorder: Fix typo

[Akira Yokosawa]

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
Hi Paul,

I don't see any smp_store_release() in P1().
It must be P0(). :-)

        Thanks, Akira
--
 memorder/memorder.tex | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/memorder/memorder.tex b/memorder/memorder.tex
index f85cfe6ca702..a0a87d0a8bca 100644
--- a/memorder/memorder.tex
+++ b/memorder/memorder.tex
@@ -2372,7 +2372,7 @@ The only code difference is that
 has an additional \co{P2()} that does an \co{smp_store_release()} to
 the \co{x2} variable that \co{P0()} releases and \co{P1()} acquires.
 The \co{exists} clause is also adjusted to exclude executions in which
-\co{P2()}'s \co{smp_store_release()} precedes that of \co{P1()}.
+\co{P2()}'s \co{smp_store_release()} precedes that of \co{P0()}.
 
 Running the litmus test in
 \cref{lst:memorder:A Release-Acquire Chain With Added Store (Ordering?)}
-- 
2.25.1

Re: [PATCH -perfbook 2/2] memorder: Fix typo

[Paul E. McKenney]

On Wed, Sep 28, 2022 at 11:30:40PM +0900, Akira Yokosawa wrote:
> Signed-off-by: Akira Yokosawa <akiyks@gmail.com>

Good catches, queued and pushed both, thank you!

							Thanx, Paul

> ---
> Hi Paul,
> 
> I don't see any smp_store_release() in P1().
> It must be P0(). :-)
> 
>         Thanks, Akira
> --
>  memorder/memorder.tex | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/memorder/memorder.tex b/memorder/memorder.tex
> index f85cfe6ca702..a0a87d0a8bca 100644
> --- a/memorder/memorder.tex
> +++ b/memorder/memorder.tex
> @@ -2372,7 +2372,7 @@ The only code difference is that
>  has an additional \co{P2()} that does an \co{smp_store_release()} to
>  the \co{x2} variable that \co{P0()} releases and \co{P1()} acquires.
>  The \co{exists} clause is also adjusted to exclude executions in which
> -\co{P2()}'s \co{smp_store_release()} precedes that of \co{P1()}.
> +\co{P2()}'s \co{smp_store_release()} precedes that of \co{P0()}.
>  
>  Running the litmus test in
>  \cref{lst:memorder:A Release-Acquire Chain With Added Store (Ordering?)}
> -- 
> 2.25.1
> 
>